Instrument approach from cockpit
13 min read

Without bustin’ your…

There’s a big difference between finding the VFR and the IFR sweet spot on an arrival. Weather doesn’t play much of a role when it’s really VFR and it plays a humongous role when it is IFR. In fact, weather determines the location of the IFR sweet spot. Sweetest of all would be when the runway pops into view at minimums with the airplane speed and configuration in perfect order.

There’s good news and bad news on IFR accidents, too. The good news is that the number of IFR accidents is down quite substantially with less than half and many in recent years as were found, say, 15 years ago. The bad news is that the amount of IFR flying is also down substantially. How much is anybody’s guess but any way you slice it the accident rate probably isn’t a lot, if any, better.

The optimist in me wants to find dramatic improvement brought about by the introduction of so much high-tech equipment in private airplanes used for IFR flying. Studies that have been done do not show advantages but the changing nature of the activity and the pilots involved make any accurate determinations elusive.

FlightAware map

One look at FlightAware shows that Cirrus SR22s are the new Bonanza or Cessna 210.

In looking at the private airplanes in the IFR system on a day with some weather around, I found only a trace of the good old days. There were but three Cessna 210s out there and five each Bonanzas 36s and Baron 58s. There were, however, 20 Cirrus SR-22s and multiple visits to FlightAware bear out the fact that the SR-22 has become the flagship of the private IFR fleet. The only other piston single that comes close is the 172 and those are likely training flights, in the system, as they should be.

This is significant because the SR-22s are mostly equipped with high-tech flight management systems and most are flown by new-breed pilots. The old-breed pilots are likely still flying the few 210s, Barons and Bonanzas found in the system.

I bring this up because orchestrating an IFR arrival and approach is a different matter with a capable flight management system than it is with charts and basic equipment. The same things have to happen but the methodology used to make them happen is not the same.

This harks back to something that I think is one of the most important elements of instrument/IFR flying. Do it all the time, on every flight. Follow the procedures and fly the approaches regardless of the weather. That way, setting up the avionics or folding the charts correctly will become second nature. The goal is to make a bad weather arrival as much like a good weather arrival as possible.

Any pilot can have a bad day but one had a terrible day when he tried to arrive at his home airport as the day was coming to an end.

Both the airplane and pilot were, well, mature. The pilot was 73 and the airplane a 1967 V35A Bonanza.

The pilot had, according to the NTSB report, about 5,000 hours in the make and model so he was no newcomer. He had a recent instrument proficiency check in the airplane. His 3rd Class medical was valid but was just issued for one year. It appears that the pilot was forthcoming on the application with medications listed for diabetes and hypertension.

The airplane had been upgraded with a Garmin glass cockpit and ADS-B capability. The autopilot was a King KFC-200 which is pretty old but a really good autopilot for this airplane. A lot of pilots thought it was the first real autopilot for prop airplanes. No mention was made of specific training with this equipment but there was that recent IPC.

The nearest weather reporting station showed 600 overcast and four miles visibility.

That all sounds good but not much good happened as the day turned into night while the pilot made unsuccessful attempts to craft an arrival into his home airport.

In fact, the pilot tried to get set up for the approach six times.

For the first approach the pilot was given vectors for the final but he flew through it and when given a vector to rejoin final he flew through it again. The controller asked if he had heard his transmissions and the pilot replied that he was “busy flying the airplane.”

On the second approach the pilot again flew through final and the controller terminated the approach. Then the pilot asked to fly the full approach, starting from outside the initial approach fix. The controller terminated that approach because the pilot was making S-turns on final.

Those were GPS approaches. For the fourth approach the controller suggested an ILS but the pilot flew through the final on this one as well so the controller cancelled the approach clearance and advised the pilot of airports in the area with weather good enough for a visual approach.

Garmin G600

That glass cockpit is amazing, but you have to know how to work the knobs and buttons.

The pilot replied with a request for another full GPS approach. The controller asked if it might help to talk to a different controller who was also a pilot and the pilot replied that he just wanted to fly the full GPS approach with no shortcuts.

There was more confusion and that approach was abandoned and for the sixth try it was to be vectors for another GPS approach. The pilot mentioned that he was “getting tired of flying this airplane” and when asked again if he’d like to go to an airport with visual conditions he said he would like to continue trying to land at this airport and “keep working until we get it.”

Shortly after that the airplane entered a rapid descending turn that continued until terrain impact.

It is clear that this pilot was unable to reach the IFR sweet spot and there are a lot of unanswered questions.

If programmed properly, the autopilot would have intercepted and flown the final approach. The report made no mention of the autopilot being inoperative so perhaps it was a matter of the pilot preferring hand-flown approaches. To many pilots, especially newer ones, that seems foolhardy but a lot of old pilots never trusted autopilots.

There is for a fact a huge difference between the autopilots of old and the latest versions. I was one of those pilots who didn’t care much for autopilots and in fact never owned anything more than a roll-control autopilot. The 182 I bought new in 1968 had no autopilot at all, nor did my 1974 Skyhawk, also bought new.

In doing pilot reports, I did fly a lot of airplanes with the best available autopilots and it was my experience with these that killed any desire to own one. In my opinion none were worth having. I think the newer autopilots are much better and I enjoyed using the Garmin GFC 700 that was paired with the G1000 in the airplanes used in my last year of flying.

I had an S-Tec roll/heading/tracking unit on my P210 for the last years I flew that airplane and I used the S-Tec a lot. It was simple, honest and reliable.

An autopilot, even a simple one, can be your best friend when you are looking for that IFR sweet spot. With my simple autopilot, I used the heading function while maneuvering for an approach and let the nav tracking function manage the localizer on an ILS or the final approach course on a GPS approach. That meant I had to manage the pitch/power/altitude part but that was easy enough to do and I felt it kept me in the flying loop throughout the approach.

With the Garmin system I programmed the navigation and the approach and used the autopilot to manage altitude and to fly the ILS or GPS with vertical guidance. To be honest, I didn’t think the flying was any easier with the full autopilot than with my simple system but it was different.

There is no shortage of pilots who will contend that approaches should be hand-flown, for the experience if nothing else. The best argument against this is that when the autopilot is flying the human pilot is a back-up and monitor. When the human pilot is flying, there is no back-up or monitor.

Weather is a critical part of any instrument approach and a pilot who understands everything about the weather for the approach is way ahead of the game.

The high-tech equipment has brought complete weather information to the cockpit but it is still up to the pilot to interpret and understand the information.

Wind shear diagram

Wind shear is not a minor footnote in a TAF.

I’ll offer an example. I think a lot of pilots see something like WS020/24045kts on a TAF right after the cloud description as not a lot more than a collection of letters and numbers. In reality this is word on something that can have a profound effect on the airplane during the descent on an approach.

Those letters and numbers mean there will be wind shear from 2,000 feet a.g.l. down to the surface and that the wind at 2,000 feet is expected to be from 240 at 45 knots. The criteria for this is a change in velocity of 30 knots from the stated altitude down to the surface so in this case the basic surface wind would be under 15 knots. I did look at some WS forecasts on a day that was ripe for the condition and some seemed to be based on less than a 30 knot change.

A changing wind does affect an airplane in flight and when descending on approach a decreasing tailwind during the descent will cause the airplane to trend high and fast where a decreasing headwind will make it trend low and slow. If a pilot know this is coming, it is a lot easier to deal with. Wind shear can also be anticipated by comparing the actual wind on the final approach course (using GPS) with the reported surface wind to get an idea about changing wind.

When descending, the changing wind can be felt as light turbulence.

The granddaddy of wind shear is found in frontal zones and around thunderstorms. If either of these is present as an approach is started, best know exactly how it might affect the airplane. Often the best answer is it won’t do anything because the pilot will be smart enough to fly the airplane to an unaffected airport for a more serene arrival.

Over the years I have noticed that aviators have a decreasing interest in the fine points of weather. When I first started working at FLYING in 1968, weather was one of our most popular subjects. I always thought that pilots felt there was an air of mystery about weather.

Now I think pilots feel there is less mystery, what with all the weather information right there on a screen in the cockpit. Trouble is, the one absolute about weather remains and will always remain unchanged: What you see is what you get. That is true regardless of what someone tells you or what you read on a screen. A METAR is nothing but a snapshot of a tiny area on or around the airport and a TAF is nothing but an educated guess about what the weather might be. The only forecasts I have seen improve over a lot of years of watching are wind aloft forecasts. They used to be terribly inaccurate, especially compared with what they are now.

Back to what you see is what you get for a moment: This is critical in the last stages of an approach. When at minimums or the missed approach point, look up and if what you are supposed to see is not there, regardless of what the reported weather might be, that is a mandate to miss the approach. If scattered clouds are reported do be aware that they can have the same effect as an overcast because slant-range visibility is all that counts and if there is a bit of scud between you and the runway the view might be blocked.

Instrument approach from cockpit

Slant range visibility is what really matters at DH.

Over many years I completed approaches when the reported weather was below minimums and missed them when it was above minimums. For example, on an approach when almost-VFR conditions were reported, I was still on top of clouds at the minimum attitude.

If the missed approach is started, too, let that be it. I recall one approach when I went into the miss and then saw enough to land visually. Trouble was, I wasn’t sure how much runway was left. I landed and it worked out but the brake store got some business because of the smokin’ stop. Lesson learned.

Finally, when starting an instrument approach it is important to be aware of the fact that from the time you leave the final approach fix to the time you either land or are established in a missed approach it is an extremely high risk form of flying, right up there with the first three minutes after takeoff. The pilot must be at the top of his game for this whether hand-flying or programming the flight management system.

If that sweet spot is never reached and a missed approach is necessary the pilot needs to be flying with the realization that this is something that happens infrequently in a flying career. That means you better have the knowledge and proficiency to fly this and get with the program promptly.

Confusion is a pilot’s worst enemy and can really come down hard after a missed approach, as in what do I do now? Before starting any approach in adverse conditions it is best to have in mind where you want to go next if the approach is missed. If this isn’t done then there’s a failing grade on what happens next, and next, and after that.

Because there is a long and tragic history of accidents on multiple approaches it is certainly not advisable to fly multiple approaches to the same airport unless something about the weather has changed for the better. The pilot at the first of the story who crashed on his sixth attempt is a good illustration of this.

There’s extra challenge in diverting to an alternate, too, because in many cases this is sort of a last chance. There might not be enough fuel in the tanks to do much other than go there, fly the perfect arrival and approach, see the runway at the sweet spot and land. Then you have to get to where you really wanted to be. The last time I did that neither of the two people with me had any cash for the $100 cab fare to where our cars were so the ride had to be on me. Being prepared also meant having cab fare or, now, an Uber account.

Richard Collins
8 replies
  1. Don Rieser
    Don Rieser says:

    Another great article from Richard Collins, although the affirmation that I am an “old breed pilot” was a bit painful. After 38 years of IFR with no autopilot, I’m convinced it is time. After spending a small fortune, by my standards, to finally equip with certified GPS navigation and ADS-B out, it may be another year or so, but perhaps there will be an STC to install a G5 with autopilot. Thanks again Richard.

  2. Mark
    Mark says:

    Excellent article – as always.
    I’m confused by just one thing. You wrote, “a decreasing tailwind during the descent will cause the airplane to trend high and fast where a decreasing headwind will make it trend low and slow. ”
    I would expect a decreasing tailwind to result in decreased groundspeed and a decreasing headwind to result in increased groundspeed.
    What am I misunderstanding? Thanks.

    • Richard Collins
      Richard Collins says:

      Airspeed is what matters, not groundspeed. Decreasing tailwind will make airspeed trend high, decreasing headwind will make it trend low. Change decreasing to increasing and the reverse is true.

  3. David Megginson
    David Megginson says:

    Good article, but I take issue with one point. When Mr Collins writes “The only other piston single that comes close is the 172 and those are likely training flights, in the system, as they should be,” I hope he’s not making that same assumption about my comparable 160hp Warrior II. I’ve used it as a serious IFR platform for 14 years now.

    Mr Collins has frequently written elsewhere that a plane has to cruise at 140kt or above to be a real cross-country plane. That’s no doubt true for people who need to fly longer distances, but a 120kt plane is perfectly reasonable for the 250–500nm flights that are typical for my needs and my family’s.

    I have New York, Boston, Quebec City, and Toronto all within an easy 1½–2½ hour flight from my home base in Ottawa (2½ hours for NYC, vs 8 hours driving); Fredericton and Sault Ste Marie at the far end of my round-trip, non-stop, round-trip range; and Washington DC and Halifax sometimes non-stop, depending on winds. That’s pretty-much everywhere I need to go, for work or play, that’s not across an ocean anyway.

    So next time you see a P28A or C172 IFR on FlightAware, please don’t automatically assume we’re just training. And keep up the great articles!

Comments are closed.